A license plate recognition system (LPRS) is a system used for identify license plates of vehicles. In this system, a vehicle passing through an n LPRS zone is photographed by a camera, and its license plate number is detected and read from a captured image of the vehicle, thereby identifying the vehicle. The license plate recognition system has been utilized in various fields, such as a speed enforcement system, a parking enforcement system, a security and surveillance system, etc.
In general license plate recognition systems, the correct recognition ratio for the license plates is in the range of 95% to 98%, and about 5% of captured license plates fails to be recognized due to poor photograph conditions, noises included in the captured images, damaged license plates, contaminated license plates, artificially veiled license plates, etc.
The misrecognized or unreadable data become a big obstacle to achieve primary objects of the license plate recognition system. For example, a city “S” in the capital area is running about 80 license plate recognition systems, and about 700,000 vehicles are captured for a day. In this case, even if the systems are operated in the optimal conditions which allow a recognition ratio of about 95%, the number of misrecognized license plates becomes about 35,000.
To identify the vehicles failing to be identified for a month, it is inevitable to check captured images of about 1,050,000 vehicles one by one by visual identification.
For the visual identification of the vehicle images, a preferable method is to input and compare the license plate numbers one by one. In this case, however, to input the license plate numbers by a known text input method using a keyboard is quite cumbersome and it takes much time.
Accordingly, a new license plate number input system capable of resolving problems described above should be proposed.
The background art of the present disclosure is disclosed in the Korean Patent Publication No. 1999-0081690 on Nov. 15, 1999.